Enterprise leaders who are struggling with squeezing true business value out of machine learning and artificial intelligence (AI) initiatives need to ask themselves an important question: Are they trying to bake bread or build the oven that bakes the bread?
The rhetorical question comes by way of one of Google’s top machine-learning experts, who recently offered up a simple analogy to explain why so many organizations are struggling with machine learning. One of the big problems is that too often hiring managers view “machine learning” as a singular specialization, says Cassie Kozyrkov, chief decision intelligence engineer at Google.
The truth is that two major disciplines are at play.
The first discipline is machine-learning research, which is focused on building the algorithms and general-purpose tools that people can use to apply machine learning in various situations. The second discipline is applied machine learning, which takes those algorithms and tools and folds them into very specific applications. The first is akin to building an oven. The second is like baking bread.
The problem is that many organizations build out the wrong teams for the job at hand because they’re hiring oven builders instead of bread bakers.
“Imagine hiring a chef to build you an oven or an electrical engineer to bake bread for you. When it comes to machine learning, that’s the kind of mistake I see businesses making over and over,” Kozyrkov says. “Leaders try to start their kitchens by hiring those folks who’ve been building microwave parts their whole lives but have never cooked a thing.”
As a result, organizations are struggling to hire the right people because they’re searching for nonexistent specialists—for example, PhD-trained AI researchers who also know how to apply machine learning in industry-specific applications to automate decisions about business processes they know nothing about. Kozyrkov says organizations should be seeking to build out interdisciplinary teams that fit the business objectives of a given machine learning project:
“If you’re innovating in recipes to sell food at scale, you need people who figure out what’s worth cooking/what the objectives are (decision makers and product managers), people who understand the suppliers and the customers (domain experts and social scientists), people who can process ingredients at scale (data engineers and analysts), people who can try many different ingredient-appliance combinations quickly to generate potential recipes (applied ML engineers), people who can check that the quality of the recipe is good enough to serve (statisticians), people who turn a potential recipe into millions of dishes served efficiently (software engineers), people who keep the interdisciplinary team on track (project/program managers), and people who ensure that your dishes stay top notch even if the delivery truck brings you a ton of potatoes instead of the rice you ordered (reliability engineers).”
Not all of these specialties need to be occupied by separate people, but the point is to ensure every role is covered, she says. To read more, check out Kozyrkov’s excellent Hackernoon Medium post.
